Babies, Bathwater, and Screening for Autism Spectrum Disorder: Comments on the USPSTF Recommendations for Autism Spectrum Disorder Screening

Coury, Daniel L. MD, FAAP

Journal of Developmental & Behavioral Pediatrics: November/December 2015 - Volume 36 - Issue 9 - p 661–663
doi: 10.1097/DBP.0000000000000227

ABSTRACT: Current guidelines for developmental screening and screening for autism spectrum disorders (ASDs) recommend screening of all children for ASD at ages 18 and 24 months. In a draft recommendation, the United States Preventive Services Task Force finds insufficient evidence to support this practice. Some of the assumptions behind these recommendations fail to consider other benefits of developmental surveillance and screening that ensue from periodic formal screening of all children. Primary care clinicians should err on the side of discovery and advocate for continued formal screening at designated intervals.

Division of Developmental/Behavioral Pediatrics, Department of Pediatrics, Nationwide Children's Hospital, The Ohio State University, Columbus, OH.

Address for reprints: Daniel L. Coury, MD, FAAP, Developmental/Behavioral Pediatrics, Department of Pediatrics, TH3D-G350, Nationwide Children's Hospital, 700 Children's Drive, Columbus, OH 43205; e-mail:

D. L. Coury is supported in part by Autism Speaks; by cooperative agreement UA3 MC 11054 through the US Department of Health and Human Services, Health Resources and Services Administration, Maternal and Child Health Research Program to the Massachusetts General Hospital; and by NIMH Grant R01MH097726-01A1.

Disclosure: D. L. Coury has received grant support from SynapDx, Stemina, HRSA, Autism Speaks, and NIMH, has served on data safety monitoring board for Neuren, is a consultant for Cognoa, and served on speaker's bureau for Abbott Labs.

The content is solely the responsibility of the author and does not necessarily represent the official views of HRSA, NIMH, or Autism Speaks. The funding sources had no role in the writing of the report and in the decision to submit the article for publication.

Received August , 2015

Accepted August , 2015

Article Outline

Developmental screening has long been held to be a fundamental part of the practice of pediatrics, and efforts to improve developmental screening have been ongoing for many decades. As methods and best practices have evolved, the use of data to support consensus or expert opinion has been used more frequently. Current guidelines for developmental screening, and screening for autism spectrum disorders (ASDs), are based on the available evidence along with consensus expert opinion. As new data become available, it is important to periodically review it as a whole and determine if these guidelines require modification.

On August 3, 2015, the United States Preventive Services Task Force (USPSTF) released its draft recommendations regarding screening for ASD. The Task Force works to improve the health of all Americans by making evidence-based recommendations about clinical preventive services, such as screenings, counseling services, and preventive medications. These recommendations are presented as a letter grade (an A, B, C, or D grade or an I statement for Insufficient Evidence) based on the strength of the evidence and the balance of benefits and harms of a preventive service. It is important to note that the recommendations apply only to people who have no signs or symptoms of the specific disease or condition under evaluation, and the recommendations address only services offered in the primary care setting or services referred by a primary care clinician.

The USPSTF draft report agrees with the current American Academy of Pediatrics screening guidelines regarding any person (family member or health professional) having any developmental concern as justification for screening for autism. Where the report is in disagreement is in its conclusion that there is insufficient evidence to support screening when there is no reported concern. The draft recommendation states:

The USPSTF concludes that the current evidence is insufficient to assess the balance of benefits and harms of screening for autism spectrum disorder (ASD) in children for whom no concerns of ASD have been raised by their parents or clinical provider.

The report confirms the existence of readily available effective screening tools and the effectiveness of early intensive interventions. Where the report finds fault is in the evidence to support screening of asymptomatic children, and it is here that many child health professionals feel the recommendation is “throwing the baby out with the bathwater.” Implicit in the recommendation are 3 assumptions: (1) a lack of parental concern means there is a lack of symptoms, or there are only mild symptoms; (2) lack of professional concern equates with a lack of symptoms, or there are only mild symptoms; and (3) there is insufficient data regarding the benefit of treating asymptomatic screen fails through early intervention. These assumptions are shallow and should be examined more carefully.

1. Regarding screening when there is no reported concern on the part of the parent.

One fallacy is the belief that all parents are aware of what normal development is and that they know for certain that their child is making appropriate developmental progress. Several studies have shown that many parents do not recognize signs of developmental delay in walking, talking, etc.1 The reasons behind this are several including lack of knowledge and cultural factors.2,32,3 As a result, caring parents may not raise any concerns of a developmental delay or possible ASD when such symptoms may be present. Choosing not to screen because of lack of family concern is not well supported by existing research.

2. Regarding screening when there is no reported concern on the part of the professional.

A second misconception is the belief that all health care professionals will recognize all developmental concerns in the limited time frame of a health supervision visit. The health supervision visit has been shown to fail to identify all children with developmental concerns because some behaviors that might be of concern are not being displayed, or because broad developmental surveillance methods do not address the child's areas of delay. The recommendation for formal developmental screening at set ages is intended to identify delays that are overlooked by surveillance. When child health professionals use only their clinical impressions rather than formal screening, estimates of children's developmental status are much less accurate.4 More recently, Gabrielsen et al5 demonstrated that brief clinical observations (10 minutes) of children with ASD by experienced professionals can fail to identify ASD. Children who had autism showed more typical behavior (89% of the time) than atypical behavior (11%) overall. Expert raters missed 39% of cases in the autism group as needing autism referrals based on these brief but highly focused observations. As a result, health care professionals may not identify concerns of autism through the health supervision visit alone; formal screening is needed.

The task force recommendation also implies that asymptomatic children identified through screening have milder symptoms and may not require or benefit from early intervention. As discussed here, screening does not solely identify patients with subtle symptoms or mild cases; it identifies symptoms that may not be displayed during the usual health supervision visit. Choosing not to screen because of lack of professional concern is not well supported by existing research.

3. From the report: “In general, children identified through screening rather than through case finding are likely to be younger and possibly less severely affected than those in study populations. It is therefore unclear whether young children with ASD detected by screening and not because of parental or teacher concern will experience similar, or any, benefit.”

The statement that children identified through screening are likely to be younger is valid and is the goal of screening for ASD or any condition—earlier identification to initiate early treatment. However, this interpretation fails to appreciate the process of referral to, and enrollment in, early intervention services. A child who shows no symptoms or fails to raise concerns by parents or health care provider, who then screens positive for autism, does not go directly to early intensive intervention. Instead, the child will then go through extra recommended steps. These range from close monitoring through a follow-up visit in a month or 2 to further evaluation by developmental specialists. After this further evaluation step, the child's developmental concerns are better assessed and at that time recommendations for treatment follow. As a result, only those children who fail screening and, on closer evaluation, clearly have developmental concerns are referred to early intervention. This in fact is the process for positive results of virtually all screening tests: a repeat of the screen or direct referral for diagnostic evaluation, followed by treatment on confirmation of the screened condition.

Screen-failed children do not receive early intervention services unless further preintervention assessment confirms concerns for ASD or other developmental conditions.

A study of autism screen-fail only children, who receive ASD services, will likely never be conducted. Similarly, all children included in clinical trials are the ones who failed some initial evaluation, potentially screening, and then had more in-depth evaluation to document the presence of autism. Thus, they do represent children who will undergo initial screening and then a diagnostic evaluation.

In making the statement above, the USPSTF suggests that screening of children where there are no concerns will only identify milder cases. Again, the studies previously cited indicate this is not correct.

The Task Force is correct in determining that there are no studies of evidence of benefit or meeting inclusion criteria of clinical outcomes of children identified with ASD through screening. What the interpretation fails to consider is the secondary benefits of this screening. There are socioeconomic,2 racial, and ethnic disparities in the identification and treatment of children with ASD.6 Routine, standardized screening for ASD has the potential to reduce disparities in age at diagnosis and possibly reduce racial and ethnic disparities surrounding access to services.6,76,7

The Task Force's conclusion is based on the above described assumptions, and their determination overlooks issues as outlined in this response. The draft recommendation wording is repeated here:

The USPSTF concludes that the current evidence is insufficient to assess the balance of benefits and harms of screening for autism spectrum disorder (ASD) in children for whom no concerns of ASD have been raised by their parents or clinical provider.

There is potential for the Task Force's findings to be misinterpreted. The wording regarding insufficient evidence is too easily interpreted as “there is no need to do this” rather than “we are not certain this is helpful.” This presents the potential to “throw the baby out with the bathwater.” The draft statement in its current wording sends a message that there is no need for screening unless concerns of ASD have been raised by the child's parents or clinical provider. The evidence cited here suggests this is an incorrect determination as it is focused more on a lack of evidence for assumption 3 above and does not consider the larger clinical situation. We do have evidence that we are already missing children with developmental and ASD concerns with our current screening processes; we should not be making this worse by stopping screening in asymptomatic children. We have evidence of effective screening strategies, evidence of effective interventions, and evidence that early identification and intervention lead to best outcomes. While we await further evidence to assess the benefits of screening asymptomatic children, it is prudent to err on the side of caution and continue to screen all children in accordance with existing guidelines.

Back to Top | Article Outline


1. King TM, Rosenberg LA, Fuddy L, et al.. Prevalence and early identification of language delays among at-risk three year olds. J Dev Behav Pediatr. 2005;26:293–303.
2. Mazurek MO, Handen BL, Wodka EL, et al.. Age at first autism spectrum disorder diagnosis: the role of birth cohort, demographic factors, and clinical features. J Dev Behav Pediatr. 2014;35:561–569.
3. Zuckerman KE, Sinche B, Cobian M, et al.. Conceptualization of autism in the Latino community and its relationship with early diagnosis. J Dev Behav Pediatr. 2014;35:522–532.
4. Werner EE, Honzik MP, Smith RS. Prediction of intelligence and achievement at ten years from twenty months pediatric and psychologic examinations. Child Dev. 1968;39:1063–1075.
5. Gabrielsen TP, Farley M, Speer L, et al.. Identifying autism in a brief observation. Pediatrics. 2015;135–e338:e330–e338.
6. Herlihy LE, Brooks B, Dumont-Mathieu T, et al.. Standardized screening facilitates timely diagnosis of autism spectrum disorders in a diverse sample of low-risk toddlers. J Dev Behav Pediatr. 2014;35:85–92.
7. Daniels AM, Mandell DS. Explaining differences in age at autism spectrum disorder diagnosis: a critical review. Autism. 2014;18:583–597.

Index terms: autism spectrum disorders; ASD; developmental screening; USPSTF; guidelines

Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.